I'm working on a curious project here. I tried searching the forum but I couldn't find any similar topics, so hopefully this isn't a duplicate, but if it is, please point me in the right direction. I've got a switching power supply from an Xbox 1.4 that is on the fritz. I just got it from someone that said it doesn't turn on. It was extremely dirty (roaches, to be exact) and I pieced it apart and ran all the electronic and plastic pieces (not mechanical or optical) through a short cycle in the dishwasher, then promptly dried it all off with compressed air. Visually, all the components are now as good as new. No leaking or bulged capacitors, no discolored ICs, no burn marks on the PCB, no blown traces - everything, visually, is absolutely perfect... well, except the annoying white "glop" all over the place from the factory...

However, the original problem still seems to remain. The power supply is almost completely dead. I've got a bit of experience with electronics - I've fixed a few things, but just the same, I've been unable to fix a large stack of things as well (that I hope one day to understand how to fix). Armed with a cheap $3 digital multimeter from Harbor Freight, I've found that the initial stages of the AC filtering seem to work fine. The large (250v/470µF) filtering capacitor charges up to 166.4 volts (line voltage is 120v/60hz). But that's where the line stops.

I checked the standby (3.3v) transformer for AC output (on the low side) while plugging the power supply in to line, and I got nothing. Zero volts. All the low side capacitors are also showing 0.00 volts. Everything is dead straight from the get-go.

I've also got another power supply that exhibits exactly the same behavior. It was working just fine until I broke apart the high and low sides to try diagnosing another power supply... with, you guessed it... that same problem as well. After re-assembling and resoldering the components, the power supply no longer functions at all. And I've got another few PC ATX power supplies in the closet that, you guessed it, do the same thing. So to be able to understand why this happens on this power supply would help me fix a LOT of things.

Armed with just a cheap digital multimeter, what can I do to, perhaps, fix this?

edit: One minute, I'll try looking up some information about the Xbox power supply for those curious.

edit: Crap, there's no technical information available. I coulda sworn I made this edit days ago though...

 

i hope you are using an isolation transformer!

 

i hope you are using an isolation transformer!

A what?

Something like this?
http://www.raphnet.net/electronique/isolation/images/isolation.jpg

 

Hm, I didn't expect I'd need to bump this.

 

Bump for any useful information...?

 

Do a Google search for "ATX bench supply" - you'll find out that PC switching power supplies require a load on +5v before they'll regulate properly.

If you've tried separating the various circuits in the PS, you may've fried the MOSFETs.

 

Well as far as the load on the supply, well... even with a load, it doesn't do anything at all. It acts completely dead - zero volts; the output filter capacitors never charged (and that would provide enough of an initial load to show *some* activity, charging the capacitors). Voltage in, nothing out.

As for that other power supply I separated, I didn't plug it back in until I soldered the pieces back together (I was only using one side for testing)... then it just did the same thing that the Xbox supply is doing: bricked. Nothing. Zombified. If it blew the MOSFET(s), wouldn't they have, perhaps, gotten hot first? Or showed a sign of damage?

Another thought I had was those awkward phototransistors that provide the feedback loop for the power supply... do those ever "go out"? Like, the LED can get too much current, or the receiving end can get messed up somehow? They seem like a sort of hackish solution...

 

A schematic of the circuit (if you can find one) and a more accurate explanation of what is going on may help.

 

Do a quick check of the rectifier diodes.